[antlr-interest] Reusing an existing Parse Tree rather than reconstructing one

[Kunal Sawlani]

Hi,
I have a parse tree which gets constructed as the user types in his input.
This seems to be working fine, but the problem is it reconstructs the tree
every time the user types in something, which is undesirable.
Is there a way to modify the existing parse tree, for better efficiency
purposes. And also, I want to maintain information in the tree whether a
node is stale, i.e, changed, from before when the user finishes typing his
input.
I was looking into the TreeWizard, but could not quite come up with a
solution.

Any ideas for this, or a work around for achieving the same would be really
appreciated.

Thanks
-- 
Kunal Sawlani